6. Project Summary/Abstract. We want to understand the molecular mechanisms that underly the generation of pathogenic autoantibodies, such as those occurring in patients with systemic lupus erythematosus. These autoantibodies are somatically hypermutated and class-switched. Somatic hypermutation (SHM) introduces mainly point-mutations in immunoglobulin (Ig) heavy and light chain V(D)J regions, thereby providing the substrate for selection by antigen of higher affinity antibody mutants. Class switch DNA recombination (CSR) replaces the expressed IgH constant (CH) region, e.g., C[unreadable], with downstream C?, Ca or Ce, thereby endowing antibodies with new biological effector functions. Both SHM and CSR would entail two sequential stages: (i) generation of DNA lesions, as initiated by activation-induced cytidine-deaminase (AID) and Ung dU glycosylase, and (ii) DNA lesion repair, as dealt with by the cell abasic site bypass and base-exicision repair (BER) (Phase 1b) or mismatch repair (MMR) (Phase 2) machineries, leading to introduction of mismatches (mutations) or doublestranded DNA breaks (DSBs), the obligatory intermediates in CSR, and their resolution. We argue here that both stages, i.e., DNA lesion and DNA repair, of SHM and CSR are dysregulated in lupus B cells. We hypothesize that the phylogenetically conserved homeodomain protein HoxC4, which, as we have shown, regulates the human IgH locus, critically induces AID expression through a conserved HoxC4/Octbinding site in the AID gene promoter. We also hypothesize that HoxC4 expression is enhanced by estrogen, and estrogen-induced upregulation of HoxC4 and AID results in increased SHM and CSR. Estrogen-mediated dysregulation of SHM and CSR would significantly contribute to the generation of autoantibodies in lupus. We further contend that translesion DNA synthesis (TLS) polymerase (pol) ? and pol ? are critical to SHM of autoantibodies, by repairing DNA lesions initiated by HoxC4-induced AID and as recruited by ubiquitinated PCNA, which mediates the "polymerase switch" from replicative high-fidelity DNA polymerases to error-prone TLS polymerases. Finally, we hypothesize that the expression of HoxC4 and AID is dysregulated in lupus and ablation of HoxC4 expression or TLS polymerases in lupus-prone mice leads to delayed onset of high-affinity and isotype-switched autoantibody production and immunopathology. To test our hypotheses, we will: (i) define the role of HoxC4 in SHM and CSR, by defining the mechanisms of HoxC4-mediated AID induction, and the enhancement of HoxC4 expression by estrogen in vitro and in vivo; (ii) address the role of TLS polymerases and their recruitment in SHM; and, finally, (iii) analyze the impact of deficiency of HoxC4 and TLS polymerases on the generation of hypermutated and class-switched autoantibodies in lupus-prone mice. These experiments will use sophisticated biochemical methods and newly engineered KO hoxC4-/-, pol?-/- , double KO pol ?-/- msh2-/-, pol ?-/- ung-/-, pol ?-/- ung-/- and pol ? mutant mice, as well as lupus-prone NZBxNZW F1 and MRL/faslpr/lpr mice deficient in HoxC4, Ung or TLS pol ? or pol ?.